Authors:
Heinze K, Reniers RLEP, Nelson B, Yung AR, Lin A, Harrison BJ, et al.
Authors notes:
Biol Psychiatry. 2015;77(11):989-96.
Keywords:
Default-mode network, Network-level, Psychosis, Structural covariance, Transition, Ultra-high risk
Abstract:
Background Investigation of aberrant large-scale brain networks offers novel insight into the role these networks play in diverse psychiatric disorders such as schizophrenia.
Although studies report altered functional brain connectivity in participants at ultra-high risk (UHR) for psychosis, it is unclear whether these alterations extend to structural brain networks.
Methods: Whole-brain structural covariance patterns of 133 participants at UHR for psychosis (51 of whom subsequently developed psychosis) and 65 healthy control (HC) subjects were studied.
Following data preprocessing (using VBM8 toolbox), the mean signal in seed regions relating to specific networks (visual, auditory, motor, speech, semantic, executive control, salience, and default-mode) were extracted, and voxel-wise analyses of covariance were conducted to compare the association between whole-brain signal and each seed region for UHR and HC individuals.
The UHR participants who transitioned to psychosis were compared with the UHR participants who did not.
Results Significantly reduced structural covariance was observed in the UHR sample compared with the HC sample for the default-mode network, and increased covariance was observed for the motor and executive control networks.
When the UHR participants who transitioned to psychosis were compared with the UHR participants who did not, aberrant structural covariance was observed in the salience, executive control, auditory, and motor networks.
Conclusions: Whole-brain structural covariance analyses revealed subtle changes of connectivity of the default-mode, executive control, salience, motor, and auditory networks in UHR individuals for psychosis.
Although we found significant differences, these are small changes and tend to reflect largely intact structural networks.